Haemorrhagic cystitis is a frequent complication of treatment with cyclophosphamide. It remains a difficult clinical problem to treat, compounded by the frailty of patient. Furthermore, the preventative measures and treatments available for CP-induced haematuria have their own benefits and disadvantages.
Haemorrhagic cystitis is a frequent complication of treatment with cyclophosphamide. It remains a difficult clinical problem especially as patients may have metastatic disease, are often elderly and may be unfit for surgery.
Cyclophosphamide (CP) is an oxazaphospharine alkylating agent that acts by cross-linking strands of DNA, thus preventing the division of cells. It is a frequently used chemotherapeutic agent for treating breast cancer, B-cell lymphoma and leukaemia. The immunosuppressive actions of CP are used in conditioning before bone-marrow transplantation, in the treatment of rheumatoid arthritis and of systemic lupus erythematosis.
Whilst the use of CP has significant benefits, acrolein, a hepatic metabolite, is associated with significant urological side-effects . Excretion of acrolein in the urine produces oedema, ulceration, haemorrhage and necrosis of the urothelium [2,3]. In addition, persistent infection with adenovirus or BK virus in individuals with bone marrow suppression might prevent mucosal healing.
Haemorrhagic cystitis (HC) is a frequently reported complication of CP therapy (2–40%) . Treatment is difficult and can often be compounded by significant comorbidities. The haemorrhage is often worsened by thrombocytopaenia (inherent in carcinomas, and induced by CP) and a transfusion-induced coagulopathy. Clinicians often become reluctant to offer advanced treatments which could exacerbate a difficult situation.
Herein we review the key preventative measures and treatment options in the management of CP-induced haematuria. Preventative measures include the use of irrigation solution , sodium-2-mercaptoethane sulphonate (Mesna) and potentially nuclear factor (NF)-κB inhibitors . The mainstay of treatment currently consists of bladder irrigation with glycine solution. This has the aim of reducing overall bladder inflammation, and preventing intravesical clot formation, which can encourage further bleeding.
Other treatments of CP-induced haematuria include sodium pentosan polysulphate (SPP), intravesical formalin, intravesical alum irrigation, hydrostatic pressure embolization, and intravesical prostaglandins.
As a prophylactic measure, the insertion of a catheter and use of continuous bladder irrigation is no longer used. However, it was more common before the introduction of new prophylactic and treatment measures. One retrospective study in 1995  compared the use of continuous bladder irrigation with no bladder irrigation in 199 patients who received cyclophosphamide ± busulfan in preparation for bone-marrow transplantation. There was a significant decrease in haematuria in patients receiving irrigation (53%) compared with the non-interventional group (23%, P < 0.004).
A more recent retrospective non-randomized study  compared patients having bladder irrigation for 36 h with those who did not. Both groups of 40 patients also received treatment with Mesna, hyperhydration and urinary alkalinization, whilst having conditioning for bone-marrow transplantation. The group receiving bladder irrigation reported improvements across several variables: a lower incidence of HC (50% vs 32%; P = 0.11); a significant reduction in the mean duration of HC (18 vs 10 days; P = 0.02); a shorter duration of hospitalization (39.6 vs 30.2 days; P < 0.001); and a lower incidence of late-onset HC (P = 0.001).
The benefit of Mesna in preventing CP-induced HC is well documented and it is used routinely in bone marrow treatment centres as a prophylactic agent. An early prospective randomized study  comparing Mesna and forced diuresis showed a significant reduction in the incidence of macroscopic haematuria in bone marrow recipients in the Mesna-treated group (61 patients, P < 0.05). There was also no specific side-effects of Mesna.
However, Mesna is limited in some regards; most notably the half-life of Mesna is 90 min, which is significantly shorter than the half-life of CP (420 min). Consequently, Mesna administration must be continuous throughout the CP treatment. A study in 1991  randomized 100 patients to receive Mesna or forced saline diuresis. The patients were undergoing bone-marrow transplant conditioning with regimens including high-dose CP. The incidence of severe haematuria was 33% in the Mesna arm and 20% in the hyperhydration arm (P = 0.31). This study showed that Mesna and hyperhydration are equally effective in preventing CP-induced HC in bone marrow transplant patients.
GLUTATHIONE AND AMIFOSTINE
Glutathione and amifostine are cytoprotective agents which act to protect normal tissues against the significant side-effects of chemotherapeutic agents. Amifostine is metabolized to an active form, WR-1065, by alkaline phosphatase, which is found on the plasma membrane. Both substances act as free-radical scavengers and provide significant protective effects to haemopoietic progenitor cells.
It has also been shown that in Swiss male mice, pre-treatment with amifostine or glutathione prevents ifosfamide-induced haematuria in a dose-dependent manner. Acrolein-induced HC was also prevented by systemic (intraperitoneal) or local (intravesicular) pretreatment with glutathione or amifostine, with the greatest protective effect seen with local amifostine treatment (2 mg/kg intravesicular; P < 0.05) .
The inflammatory process in CP-induced haematuria is mediated by several cytokines, including TNF-α, interleukin-1 and cyclooxygenase-2. It is believed that transcription factors can up-regulate or down regulate the production of these mediators. One such transcription factor is NF-κB and so NF-κB inhibitors such as sesquiterpene lactone parthenolide might inhibit inflammation.
A recent study  showed that intravesical pretreatment with parthenolide inhibited bladder inflammation and overactivity in rats, whilst in humans parthenolide has also been shown to suppress the translocation and translation of NF-κB, and reduce cyclooxygenase-2 and interleukin-1 production. As a consequence, NF-κB inhibitors might be used for preventing CP-induced haematuria.
Oral SPP has been shown to be effective as an agent in managing interstitial cystitis . A recent study at the authors’ institution  also showed that SPP is an effective agent in the treatment of CP-induced haematuria and haematuria induced by pelvic radiotherapy. In 31 of 51 patients the haematuria stopped and 10 were able to discontinue treatment. The remaining 20 patients died from their malignancy before adequate evaluation of SPP therapy. The regimen suggested by that report, and others, is 100 mg of SPP three times daily for 1–8 weeks. Bladder washout and irrigation are continued as necessary. Once control is achieved, SPP is continued for 1 year and then gradually reduced.
SPP might exert its effects in several different ways. These include the replacement of surface glycosaminoglycans which might have been lost due to CP treatment; consequently, the bladder is then less susceptible to bacterial infection. Another postulated mechanism is the inhibition of mast cell degranulation, resulting in a decrease in the production of NF-κB, resulting a dampening of the immune response [13,14]. Oral SPP has been shown not to have a detectable anticoagulant activity, is safe and not toxic . SPP has been associated with thrombocytopenia and thrombosis , the mechanism of which is thought to be similar to that associated with a heparin-induced thrombocytopenia.
Formalin is used mainly as a histological fixative but its properties extend to its use as a preservative, disinfectant and an embalming agent. It was first used as an intravesical agent in 1969 and has subsequently been used successfully to treat severe CP-induced haematuria, as well as haematuria secondary to carcinoma and radiotherapy [16–18]. Formalin is thought to cause the precipitation of cellular proteins in the bladder mucosa. This has the effect of occluding telangiectatic vessels and capillaries . If formalin is absorbed into the systemic circulation it is modified into formic acid, which results in the inhibition of anaerobic glycolysis, hexokinase and cholinesterase .
Formalin is a 37–41% solution of formaldehyde and is diluted when used as an intravesical agent. Formalin can be used in varying concentrations but the principle is to use the smallest possible concentration for the shortest possible time . Formalin can be administered with a general anaesthetic or under spinal anaesthesia, ensuring that the patient is in a reverse Trendelenburg position. The regimen advocated by Fair  involved instilling a 1% solution of formalin for 10 min, followed by a washout of the bladder with distilled water. This schedule was found to be effective in a trial with 14 patients. Other regimens have used differing concentrations for varying lengths of time. Unfortunately the higher concentrations and higher duration of treatment have been associated with several side-effects, ranging from ureteric fibrosis, anuria and incontinence, to acute tubular necrosis leading to renal failure and death [18,22–24].
One of the more common side-effects of formalin therapy is the development of bladder fibrosis with subsequent urinary frequency. Moreover, formalin might be responsible for triggering bladder spasms in the neurologically intact bladder, through sensory stimulation of the bladder mucosa. In addition, narrowing and rigidity of the ureteric orifices can occur, leading to potential hydronephrosis and reflux. Thus, repeated instillations could lead to the reflux of formalin into the kidneys. As a consequence, a cystogram is essential to exclude ureteric reflux before administering formalin.
The effectiveness of formalin treatment is not in doubt, with a high success rate of 80–92%[16,17]. However, the risks of treatment, especially with high concentrations, are significant. Thus, it has been suggested that it should be used as a treatment when others have failed, and in individuals where life-long symptoms will not be a significant issue.
The micro-anatomical and pathophysiological benefits of hyperbaric oxygen have been well documented. Of these, most notable is the reduction of inflammation brought about by several mechanisms, including reduced production of immunity and inflammatory cytokines. Hyperbaric oxygen also stimulates wound repair, increased fibroblast production and collagen formation and capillary angiogenesis [25–29]. This in turn creates the correct circumstances for an ulcerated bleeding point to heal.
Hyperbaric oxygen therapy is delivered by inhalation of 100% oxygen in a hyperbaric chamber with a pressure typically 1.4 to 3 times that of ambient . Each treatment is for 1 h per exposure and normobaric oxygen is given between treatments. Treatments are given with increasing intervals of normal air between them, to avoid pulmonary oxygen toxicity . The haemoglobin becomes fully saturated and the oxygen dissolved in the plasma increases, which in turn leads to the physiological changes described above. To have this treatment the patient must be fit enough to walk into the chamber and sit for the duration. In practice, this is difficult when the patient is sick and having bladder irrigation.
Hyperbaric oxygen therapy has been shown to be effective for CP-induced haematuria. However, much of the research has focused on the successful treatment of radiation-induced haematuria and not CP-induced haematuria . A study in Sprague-Dawley rats showed that hyperbaric oxygen was effective in treating HC induced by CP, but it was not effective as a preventative measure .
INTRAVESICAL ALUM IRRIGATION
Alum irrigation was first reported in 1982 in the treatment of six patients and has been discussed in several reports . Alum is thought to act by the inducing protein precipitation at the cell surface, and in interstitial spaces. As a result there is decreased capillary permeability, contraction of the intercellular space, vasoconstriction, hardening of the capillary endothelium and a reduction in oedema, inflammation and exudate .
A 1% aluminium potassium sulphate solution, made by dissolving potash in sterile distilled water, is administered in a continuous intravesical lavage of normal saline fluid. This can be inserted into the bladder with a whistle-tipped catheter under local anaesthesia, thus avoiding the need for a general anaesthetic.
Alum instillation has been found to be effective in the treatment of HC caused by TCC, chemotherapy and radiotherapy . In early work, eight patients aged 57–82 years and with severe vesical bleeding had irrigation for a mean of 3 days; the bleeding stopped in all of them within this time .
However, in another report of five patients who developed CP-induced haematuria, despite intravenous hydration and diuretics, only one treatment with alum was successful. In addition, large hardened clots of blood and alum precipitate necessitated the use of open cystotomy for evacuation.
Alum irrigation has limited use in children with HC. The provisos that a lower concentration of irrigating solution is used, that serum aluminium levels are monitored regularly, and other investigations are repeated frequently, limits its effectiveness . This is due to the risk of aluminium toxicity, which can lead to numerous symptoms including suprapubic pain and bladder spasms, related to the acidity of the irrigation fluid. More profound problems include respiratory depression, renal failure, ataxic seizures and even dementia. Whilst these have not been reported with the use of alum treatment in adults they remain a concern in children, and hence the necessary precautions in this age-group . The administration of alum irrigation is also limited by the fact that most hospital pharmacies might not stock this product and so obtaining alum might not be possible.
The vasculature of the bladder is derived from the internal iliac arteries, and attempts to control intractable haematuria by using embolization have had some success. It was first reported in 1974 by Hald and Mygiand  and a subsequent review of reports showed successful control of severe haematuria in 32 of 35 patients (92%) .
The internal iliac artery and its main branches can be catheterized via the ipsilateral or contralateral femoral artery, or from the left axillary artery. Embolization can then be applied by the use of Gelfoam or coils [41,42]. It has also been done using a blood clot and isobutyl-2-cyanoacrylate [43,44]. Selective embolization works on the same principle as embolization, but there is a greater degree of control of specific blood vessels.
More recently, superselective embolization has been used to treat severe intractable haematuria. By using a microcatheter (2.5 F) instead of an angiographic catheter (4 F), together with a guidewire (≈1 mm) and using a polyvinyl acetate embolization suspension in the bladder arteries, a centre in Rome  reported that it is a feasible technique. In addition, it was reported that superselective embolization is associated with a lower recurrence rate of haematuria and a lower degree of postembolization gluteal pain, claudication or tissue necrosis [45,46].
However, the efficacy of embolization is variable. In a short series of seven patients who had embolization of the internal iliac artery, with a follow-up was 6–12 months , the haematuria was controlled in four patients ‘permanently’, while three had further episodes of haematuria necessitating admission. Two of these patients were managed conservatively, whilst the third required a repeated course of embolization.
In the authors’ limited experience, embolization has never worked. The blood supply of the bladder is so extensive that adequate devascularization is difficult.
Several case reviews and case reports have suggested a potential role for prostaglandins in the treatment of severe HC, but the exact mechanism of action is unclear. It might act through several different mechanisms. These include the induction of haemostasis with platelet aggregation or by causing vasoconstriction of vessels in the mucosa and submucosa, through membrane stabilization. Topical prostaglandin E2 (dinoprostone) has been shown to be effective in controlling severe chronic HC after CP therapy . In addition, another prostaglandin analogue, prostaglandin-F2α (carboprost tromethamine) has been reported to be effective in the treatment of this HC. The occurrence of spasms as a side-effect shows their potent effect on smooth muscle in the bladder wall, and a similar effect on vascular smooth muscle might decrease blood flow locally. Promotion of platelet aggregation by prostaglandin-E2 analogues or low concentrations of carboprost might also reduce bleeding. In addition, prostaglandins are thought to have a protective effect against mucosal injury mediated by oxidants.
HELMSTEIN’S BLADDER HYDRODISTENSION
The concept of bladder hydrodistension was first reported in 1966 . It was initially described for the treatment of bladder carcinomas but was adapted for use in treatment in radiation-induced haematuria. It has never been used in the treatment of CP-induced haematuria and is probably only of historical and theoretical interest.
There are several available preventative measures and treatments available for CP-induced haematuria. Each one has benefits and disadvantages there is no ideal management option. This is worsened by the fact that patients could have metastatic disease, or be elderly and unfit for surgery. Of all the preventative options that are available Mesna appears to have the greatest effectiveness in the treatment of CP-induced HC. Mesna is used routinely in bone marrow treatment centres as a prophylactic agent. However, it is limited by its shorter half-life (90 min) than that of CP (7 h). Oral SPP and supportive measures until it is effective should always be the first line treatment. All of the other treatments have major side-effects or are difficult to instigate, or both.